I have been working lately on “Outpost”, a video adventure for the ZX Spectrum 48K. It is a bit metroidvania in some ways, because some objects unlock abilities that allow access to new areas. I guess we could say it is a bit more like the Dizzy games, if we think about games from back in the day.
I wasn’t working on this game, but I got a bit stuck in the project I was working on, so I guess having a few WIPs can also be an advantage. I really wanted to release this game back in 2023, but it didn’t happen. Looking at the repository, the first commit was in April 2021, and I think the first time I mentioned it in these pages was in one of my gamedev recaps, in June 2021. I always had that thing of not wanting to discuss a game until I was sure I would finish it, and you can see why!
This game is my attempt to make a “big” 48K game, using 100% my own code. Previously, when I worked on 48K games, I was using the sp1.lib shipped with Z88DK. That library is very good, but it uses a lot of memory. And that’s why my Castaway is a tad short.
Ignore the framerate, it is not accurate
When I worked on Brick Rick: Graveyard Shift, I wrote my own library of functions, and that included a lot of code that I didn’t end up using because Brick Rick was a 128K game. I was never happy with Castaway, so I decided to make a good 48K game –we shall see when it is finished–. I also doubled down on technology, by exploring the ZX Spectrum floating bus, and then it all went a bit cold in the level design part of it.
I think one of the reasons was because the meta-tile engine –the map is built with “tiles” that are really a “group of tiles”– was too complicated to use, and it made designing screens a bit cumbersome. So I decided to port the engine I used in The Heart of Salamanderland, and started converting all the screens I had designed.
Once I had converted most of the screens, I did some brainstorming with my sons to come up with the objects –and their uses– and address some gaps I had in my plan, and here we are: 53 screens, 3859 bytes total (72.81 bytes on average per screen). I have now more than half of the planned screens, and only a couple of things need to be coded, so it looks like the game will be done in 2026!
I’m thinking about the loading screen. When I have something, either drawn by me or perhaps with an actual artist contributing one, I will set a placeholder website for the game with further information.